The SAR Remote Sensing Image Change Detection Based On Image Fusion And Denoising Algorithm In NSCT Domain

In recent years, change detection technology has been widely used in many fields.Such as dynamic video surveillance, medical diagnosis, natural disaster monitoring and oasis cover monitoring. Because of Synthetic Aperture Radar(SAR) is not afraid bad weather such as rain and snow, also can cover through the surface of the vegetation, not easily affected by light intensity and other advantages. It has been widely used in the research of change detection.The technology of SAR remote sensing image change detection is based on the analysis of the same location with different moment in the remote sensing image data, which can be used to identify the period of the ground cover change process.This paper mainly studies the image processing algorithm based on NSCT domain of SAR remote sensing image change detection algorithm, and apply it in oasis/vegetation coverage on the remote sensing image.1、This papper analyzed the influence about SAR remote sensing image change detection results with different image fusion algorithm based on several difference-images in the NSCT domain. And analyzed the influence about SAR remote sensing image change detection results with different image fusion algorithm based on several fusion rules in the NSCT domain.Based on the above analysis,an change detection algorithm is proposed in this paper. The change detection technique is based on NSCT domain with a new image fusion rule and FLICM clustering algorithm. After that, we add denoising algorithm to the mentioned new algorithm for change detection process, thus further weakened the SAR coherent noise interference of change detection results.The proposed algorithm is applied to the real remote sensing data sets. According to comparative and analysis we found that the proposed algorithmh has high change detection accuracy.2、Analyzed the denoising algorithm in the use of remote sensing image change detection, and introducing the hidden markov tree model on the basis ofdecomposition reconstruction in the NSCT domain.Thus constructs the NSCT-HMT denoising model, finally we combined the denoising model with FLICM clustering algorithm and applied it to the SAR remote sensing image change detection.Experimental results show that the proposed SAR remote sensing image change detection algorithm based on NSCT-HMT denoising model can obtain more change detection accuracy in simulation remote sensing data set and real remote sensing data sets.3 、 Finally, taking on the southern fringe of Gurbantunggut Desert Oasis in Xinjiang Uygur Autonomous Region of China as an example, using the proposed method based on NSCT-HMT denoising model algorithm for vegetation coverage change detection.After a series of image preprocessing, image denoising, and change detection to the original remote sensing data of image, also obtained the ideal change detection result.The experiment results have great significance to the research of land use, desertification control and vegetation monitoring research.